Availability, ease of use, relative low cost, accuracy and reproducibility have made dual energy x-ray absorptiometry (DXA) the most widely used technique worldwide to obtain bone measurements for both research and clinical purposes in adult populations. However, errors related to growth and maturity greatly diminish the value of DXA bone measurements in pediatric populations. Several investigators have found that osteoporosis in children is frequently misdiagnosed secondary to results from DXA scans. In this regard, the official position of the International Society for Clinical Densitometry states its recent recommendations that subjects <20 years of age should not be diagnosed with osteoporosis based on DXA criteria. Nevertheless, the increased awareness that osteoporosis has its antecedents in childhood and the demand for examinations of bone acquisition and response to therapy in pediatric patient populations stress the urgent need to optimize DXA measurements for children. The specific aims of this proposal are to improve the accuracy of DXA measurements by assessing and correcting for the influence of bone shape, bone size, and fat and muscle volume and distribution on DXA measurements in the axial skeleton of children and young adults. Vertebral bone density will be determined by computed tomography (CT), and the volumes of the lumbar vertebral bodies, abdominal fat and paraspinous and abdominal musculature in children and adults will be assessed by magnetic resonance imaging (MRI). Based on the data collected from MRI and CT, we will be able to develop and validate sex- and age-specific formulas that will correct for the influence that bone size and body composition has on DXA measurements. The incorporation of these correction factors into current and future DXA equipment will greatly improve the accuracy of their bone measurements. In addition, the relationships between outcomes from bone age and physical examinations (Tanner stage, height, sitting height, weight, BMI, abdominal diameter, waist circumference, hip circumference, waist/hip ratio and skin fold measurements) and DXA values will be examined in an attempt to establish which clinical parameter(s) can be used in the optimization of DXA bone measurements.